BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Drupal//recurring_events_ical//2.0//EN
BEGIN:VEVENT
UID:23ce581e-8b03-4ac6-865e-b8bbc9879045@support.access-ci.org
DTSTAMP:20260310T115851Z
DTSTART:20260410T150000Z
DTEND:20260410T170000Z
SUMMARY:Introduction to Scientific Machine Learning
DESCRIPTION:Scientific Machine Learning (SciML) is revolutionizing how we m
 odel complex physical systems by blending classical numerical analysis wit
 h the flexibility of deep learning. This workshop, jointly sponsored by NC
 SA and Illinois Computes, offers a deep dive into the core principles of S
 ciML, specifically focusing on the power of Neural Operators. Using DeepO
 Net as our primary example, we will demonstrate how to move beyond simple 
 function approximation toward learning the underlying operators of partial
  differential equations (PDEs).Topics covered in the workshop include: Th
 e SciML Landscape: An overview of deep learning fundamentals and the trans
 ition from conventional surrogate models to mesh-independent neural operat
 ors. DeepONet in Practice: A practical introduction to DeepONet architect
 ures based on multilayer perceptrons (MLPs), with demonstrations on proble
 ms such as anti-derivatives and heat conduction. Advanced Architectures: 
 Strategies for scaling SciML models using CNN-based branch networks to cap
 ture complex spatial dependencies in thermal systems. Future Frontiers: A
  discussion of recent advances in operator learning and the growing role o
 f SciML in modern engineering and physical sciences.Prerequisites: No prio
 r experience with SciML is required, though familiarity with basic machine
  learning concepts and Python is recommended.Instructor Bio: Diab Abueidd
 a is a Research Scientist at the National Center for Supercomputing Applic
 ations (NCSA) at the University of Illinois Urbana-Champaign. His research
  spans deep learning, artificial intelligence, solid mechanics, multiphysi
 cs, and additive manufacturing, with a particular focus on applying scient
 ific machine learning to computational engineering problems. Qibang Liu i
 s a Research Scientist at the National Center for Supercomputing Applicati
 ons (NCSA) at the University of Illinois Urbana-Champaign. His research in
 terests span scientific machine learning, AI-aided engineering, multi-phys
 ics simulation, computational mechanics, peridynamics, and advanced manufa
 cturing.Hands-on participation: The workshop will use Google Colab for ha
 nds-on demonstrations.Register by April 8, 2026
URL:https://support.access-ci.org/events/8980
END:VEVENT
END:VCALENDAR